Semi-Active Control Of Lateral Vehicle Suspension Systems Base On Neural Network

In the high speed train, the train lateral ride quality greatly affects the safety of the train operation and the passengers ride comfort, and how to improve the train lateral ride quality is a serious problem. The train vehicle suspension system is divided into active suspension system and passive suspension system, and Semi-active suspension system is a special form of active suspension system. When the train is running, the parameters of the passive suspension system can not be adjusted, thus the train lateral ride quality turns worse; the parameters of the active suspension system can be adjusted timely, but the cost of the active suspension is so expensive that it could not be used in the reality; the cost of the semi-active suspension system is cheap, so it is an effective way to improve the train lateral ride quality. This paper applies neural network semi-active suspension control system to improve the train lateral ride quality. This is the important content of this paper.Research on the advantages and disadvantages of the Lateral semi-active suspension dynamics model in different degrees of freedom of the train, and select the 17 degrees of freedom which can better show the mode of the running train as the controlled object model. Then, use the MATLAB to build its SIMULINK model.Research on the numerical simulation of track irregularity, and Simulate the German high speed, low interference track spectrum with Precision.During the simulation of the secondary lateral suspensions for the high speed passenger vehicle with no-bolster bogie, the effective using neural network predictive control to improve the ride comfort are studied. The results of simulation show that, compared to the passive suspension system, using skyhook damping, the RMS values of lateral acceleration car body can be reduced about 40%, maximum values of lateral acceleration can be reduced 30% to 40%, the ride comfort index can be reduced about 14%.so the neural network predictive control using in the semi-active suspension system is effective.During the simulation of the secondary lateral suspensions for the high speed passenger vehicle with no-bolster bogie on different speed, the robust of neural network predictive control are studied. The results of simulation show that, with the increased speed, the ride comfort index of both passive suspension system and semi-active suspension system are increased, but the ride comfort index of semi-active suspension system still reduced 14% to 15%. So the neural network predictive control is robust.The research in this paper has proved that, it is effective using neural network predictive control to improve the ride comfort for high speed train.